@event4u/agent-config 9.3.0 → 9.4.0

package/docs/command-flows.md

@@ -7,7 +7,7 @@
 
 > Every command is classified in `src/flows/surface-map.yaml` (coverage enforced by `scripts/lint_command_flow_coverage.py`). The four user-work flows carry rich definitions in `src/flows/<flow>.yaml`.
 
-**189 commands** across **10 surfaces** (4 user-work flows + 6 platform/domain surfaces).
+**190 commands** across **10 surfaces** (4 user-work flows + 6 platform/domain surfaces).
 
 ## User-work flows
 
@@ -170,6 +170,7 @@ _The system/admin surface (NOT a flow): memory · analytics · governance · con
 - `context` — Context orchestrator — routes to create, refactor
 - `context-create` — Analyze a codebase area and create a structured context document
 - `context-refactor` — Analyze, update, and extend an existing context document
+- `contribution-precheck` — Contributor self-service precheck — run the PR-relevant lint subset (skill linter, originality gate, frontm…
 - `cost` — Cost orchestrator — routes to report (session token cost + budget ladder) and profile (change the rule_load…
 - `cost-profile` — Change the rule_loading_tier in .agent-settings.yml — shows each profile's meaning and applies the selection
 - `cost-report` — Capture token cost from the active Claude Code session, append to the local sessions store, and surface the…

package/docs/comparison.yaml

@@ -33,3 +33,13 @@ rows:
     our_evidence: "tests/scripts/witness/check_refs_relative_gap.test.ts"
     their_evidence: "Published known-limitation surfaces backed by reproducing tests are not a standard artifact in this category."
     checkable: true
+
+  - claim: "A 30-second single-file wedge exists — one read-only subagent, one command, verdict with file:line — and its promise is a scoped, published eval, not a feature list."
+    our_evidence: "internal/bench/orchestration/pv-a3-results.md#token_delta_provenance: measured"
+    their_evidence: "Category entry points typically install the full platform (profiles, packs, runtime) before the first felt win; single-artifact entries with a published eval behind the promise are not a standard artifact."
+    checkable: true
+
+  - claim: "The persona-identity question was tested and published as an honest null (identity swap ~zero effect, provider choice ~15x larger) instead of being shipped as theater."
+    our_evidence: "internal/bench/reports/persona-placebo.json#honest-null"
+    their_evidence: "Named-persona prompts are shipped as a feature across the category without a published A/B separating identity from method."
+    checkable: true

package/docs/contracts/adoption-signal-floor.md

@@ -84,6 +84,23 @@ must clear before the code lands:
 Until all four gates are written and reviewed, consumer-side
 telemetry is **forbidden** even as a discussion item in roadmaps.
 
+## Extraction demand gate — lint_originality standalone probe
+
+Added 2026-07-20 (road-to-originality-gate-and-contributor-funnel Phase 3,
+council 2026-07-20: extraction is demand-gated; the floor IS the demand gate,
+applied post-probe). Uses **allowed public signals only** — inbound GitHub
+issues/PRs and their authors; no consumer telemetry, consistent with the Iron
+Law above.
+
+| Field | Value |
+|---|---|
+| **Probe surface** | [`docs/anti-reskin-gate.md`](../anti-reskin-gate.md) — run-from-clone recipe + signal instructions |
+| **Signal definition** | A distinct external (non-maintainer) GitHub account opening an issue titled `anti-reskin gate: standalone request`, or referencing `lint_originality` in an inbound issue/PR about their own catalog |
+| **Floor** | ≥ 3 distinct external signals |
+| **Window** | 90 days from the probe page landing on `main` |
+| **Floor met** | Extract `lint_originality` as a standalone zero-config npx package (`--changed` mode, no repo coupling, regression suite included); npm publish remains a maintainer decision (name, scope, moment) |
+| **Floor missed** | Extraction **cancelled** — recorded via decision-record; not relitigated without new evidence |
+
 ## Re-audit cadence
 
 Re-audit on each of:

package/docs/contracts/ai-council-config.md

@@ -679,6 +679,31 @@ regression breaks `task consistency`, never the live council.
 `parse_corpus_strict` (Markdown). It runs **before** `task sync`
 rebuilds the lockfile, so it must read whatever the user just edited.
 
+## `model_downgrade` — auto-tiering + the A1↔A3 cache coupling
+
+Since the A3 slice of `road-to-api-cost-optimization` (2026-07-20):
+
+```yaml
+model_downgrade:
+  enabled: true        # default; false disables the size-fit gate entirely
+  auto_apply: true     # DEFAULT since A3 — auto-downgrade is opt-OUT
+  model_tier_override: # per-run escape hatch: member -> pinned model id
+    anthropic: claude-sonnet-4-5
+```
+
+- Small / low-complexity artefacts (per `necessity.classify_size_fit`,
+  character-based) auto-downgrade one ladder rung; the `debate` lens never
+  downgrades.
+- **Cache coupling:** a downgrade is applied only when the model saving beats
+  the forfeited model-scoped prompt-cache reads
+  (`pricing.downgrade_coupling`: `downgrade_savings > lost_cache_savings`,
+  expected reads = rounds − 1). One-shot paths (the low-impact fast-path)
+  have no reads to lose and downgrade to the CHEAPEST ladder rung outright.
+- `model_tier_override` pins a member to a model for the run — classifier and
+  coupling both skipped. Set it per artefact when a specific tier is needed.
+- Requires `model_ladder` on the member (smallest → largest, must include the
+  active `model`).
+
 ## `api_key_ref` forms
 
 Exactly two forms. Raw keys in the yml are a hard validation error.

package/docs/contracts/mcp-tool-inventory.md

@@ -14,22 +14,22 @@ keep-beta-until: 2026-08-14
 
 ## Summary
 
-- **Total tools:** 28
+- **Total tools:** 31
 - **By transport:** stdio=19
-- **By side-effect:** fs-write=8, ro=17, shell=3
-- **Discovery-only stubs (no implementation):** 9
+- **By side-effect:** fs-write=8, ro=20, shell=3
+- **Discovery-only stubs (no implementation):** 12
 
 ## Tools
 
 | Tool | Side-effect | Transports | Catalog | Handler |
 |---|---|---|---|---|
-| `lint_skills` | `ro` | stdio | [`consumer_tool_catalog.json:7`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L7) | [`tools.ts:1200`](../../src/scripts/mcp_server/tools.ts#L1200) |
-| `chat_history_append` | `fs-write` | stdio | [`consumer_tool_catalog.json:24`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L24) | [`tools.ts:1228`](../../src/scripts/mcp_server/tools.ts#L1228) |
-| `chat_history_read` | `ro` | stdio | [`consumer_tool_catalog.json:43`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L43) | [`tools.ts:1289`](../../src/scripts/mcp_server/tools.ts#L1289) |
-| `memory_lookup` | `ro` | stdio | [`consumer_tool_catalog.json:63`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L63) | [`tools.ts:1365`](../../src/scripts/mcp_server/tools.ts#L1365) |
-| `memory_get` | `ro` | stdio | [`consumer_tool_catalog.json:81`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L81) | [`tools.ts:1432`](../../src/scripts/mcp_server/tools.ts#L1432) |
-| `memory_signal` | `fs-write` | stdio | [`consumer_tool_catalog.json:95`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L95) | [`tools.ts:1541`](../../src/scripts/mcp_server/tools.ts#L1541) |
-| `memory_status` | `ro` | stdio | [`consumer_tool_catalog.json:111`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L111) | [`tools.ts:1456`](../../src/scripts/mcp_server/tools.ts#L1456) |
+| `lint_skills` | `ro` | stdio | [`consumer_tool_catalog.json:7`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L7) | [`tools.ts:1198`](../../src/scripts/mcp_server/tools.ts#L1198) |
+| `chat_history_append` | `fs-write` | stdio | [`consumer_tool_catalog.json:24`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L24) | [`tools.ts:1226`](../../src/scripts/mcp_server/tools.ts#L1226) |
+| `chat_history_read` | `ro` | stdio | [`consumer_tool_catalog.json:43`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L43) | [`tools.ts:1287`](../../src/scripts/mcp_server/tools.ts#L1287) |
+| `memory_lookup` | `ro` | stdio | [`consumer_tool_catalog.json:63`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L63) | [`tools.ts:1363`](../../src/scripts/mcp_server/tools.ts#L1363) |
+| `memory_get` | `ro` | stdio | [`consumer_tool_catalog.json:81`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L81) | [`tools.ts:1430`](../../src/scripts/mcp_server/tools.ts#L1430) |
+| `memory_signal` | `fs-write` | stdio | [`consumer_tool_catalog.json:95`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L95) | [`tools.ts:1539`](../../src/scripts/mcp_server/tools.ts#L1539) |
+| `memory_status` | `ro` | stdio | [`consumer_tool_catalog.json:111`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L111) | [`tools.ts:1454`](../../src/scripts/mcp_server/tools.ts#L1454) |
 | `skill_trigger_eval` | `ro` | _(stub)_ | [`consumer_tool_catalog.json:118`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L118) | _stub-only_ |
 | `suggest_command` | `ro` | _(stub)_ | [`consumer_tool_catalog.json:134`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L134) | _stub-only_ |
 | `suggest_skill_for_task` | `ro` | _(stub)_ | [`consumer_tool_catalog.json:149`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L149) | _stub-only_ |
@@ -37,20 +37,23 @@ keep-beta-until: 2026-08-14
 | `update_form_request_messages` | `fs-write` | _(stub)_ | [`consumer_tool_catalog.json:178`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L178) | _stub-only_ |
 | `sync_gitignore` | `fs-write` | _(stub)_ | [`consumer_tool_catalog.json:193`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L193) | _stub-only_ |
 | `sync_agent_settings` | `fs-write` | _(stub)_ | [`consumer_tool_catalog.json:206`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L206) | _stub-only_ |
-| `run_tests` | `shell` | stdio | [`consumer_tool_catalog.json:220`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L220) | [`tools.ts:1716`](../../src/scripts/mcp_server/tools.ts#L1716) |
+| `run_tests` | `shell` | stdio | [`consumer_tool_catalog.json:220`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L220) | [`tools.ts:1714`](../../src/scripts/mcp_server/tools.ts#L1714) |
 | `run_quality_checks` | `shell` | _(stub)_ | [`consumer_tool_catalog.json:234`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L234) | _stub-only_ |
-| `list_skills` | `ro` | stdio | [`consumer_tool_catalog.json:247`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L247) | [`tools.ts:1470`](../../src/scripts/mcp_server/tools.ts#L1470) |
-| `list_commands` | `ro` | stdio | [`consumer_tool_catalog.json:254`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L254) | [`tools.ts:1485`](../../src/scripts/mcp_server/tools.ts#L1485) |
-| `list_rules` | `ro` | stdio | [`consumer_tool_catalog.json:261`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L261) | [`tools.ts:1500`](../../src/scripts/mcp_server/tools.ts#L1500) |
+| `list_skills` | `ro` | stdio | [`consumer_tool_catalog.json:247`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L247) | [`tools.ts:1468`](../../src/scripts/mcp_server/tools.ts#L1468) |
+| `list_commands` | `ro` | stdio | [`consumer_tool_catalog.json:254`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L254) | [`tools.ts:1483`](../../src/scripts/mcp_server/tools.ts#L1483) |
+| `list_rules` | `ro` | stdio | [`consumer_tool_catalog.json:261`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L261) | [`tools.ts:1498`](../../src/scripts/mcp_server/tools.ts#L1498) |
 | `compile_router` | `shell` | _(stub)_ | [`consumer_tool_catalog.json:268`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L268) | _stub-only_ |
-| `read_resource_body` | `ro` | stdio | [`consumer_tool_catalog.json:281`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L281) | [`tools.ts:1516`](../../src/scripts/mcp_server/tools.ts#L1516) |
-| `roadmap_progress` | `fs-write` | stdio | [`consumer_tool_catalog.json:295`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L295) | [`tools.ts:1575`](../../src/scripts/mcp_server/tools.ts#L1575) |
-| `roadmap_archive` | `fs-write` | stdio | [`consumer_tool_catalog.json:308`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L308) | [`tools.ts:1598`](../../src/scripts/mcp_server/tools.ts#L1598) |
-| `capabilities_index` | `fs-write` | stdio | [`consumer_tool_catalog.json:315`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L315) | [`tools.ts:1614`](../../src/scripts/mcp_server/tools.ts#L1614) |
-| `doctor_report` | `ro` | stdio | [`consumer_tool_catalog.json:328`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L328) | [`tools.ts:1637`](../../src/scripts/mcp_server/tools.ts#L1637) |
-| `conformance_check` | `ro` | stdio | [`consumer_tool_catalog.json:335`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L335) | [`tools.ts:1651`](../../src/scripts/mcp_server/tools.ts#L1651) |
-| `telemetry_report` | `ro` | stdio | [`consumer_tool_catalog.json:342`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L342) | [`tools.ts:1665`](../../src/scripts/mcp_server/tools.ts#L1665) |
-| `council_estimate` | `ro` | stdio | [`consumer_tool_catalog.json:355`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L355) | [`tools.ts:1687`](../../src/scripts/mcp_server/tools.ts#L1687) |
+| `read_resource_body` | `ro` | stdio | [`consumer_tool_catalog.json:281`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L281) | [`tools.ts:1514`](../../src/scripts/mcp_server/tools.ts#L1514) |
+| `roadmap_progress` | `fs-write` | stdio | [`consumer_tool_catalog.json:295`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L295) | [`tools.ts:1573`](../../src/scripts/mcp_server/tools.ts#L1573) |
+| `roadmap_archive` | `fs-write` | stdio | [`consumer_tool_catalog.json:308`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L308) | [`tools.ts:1596`](../../src/scripts/mcp_server/tools.ts#L1596) |
+| `capabilities_index` | `fs-write` | stdio | [`consumer_tool_catalog.json:315`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L315) | [`tools.ts:1612`](../../src/scripts/mcp_server/tools.ts#L1612) |
+| `doctor_report` | `ro` | stdio | [`consumer_tool_catalog.json:328`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L328) | [`tools.ts:1635`](../../src/scripts/mcp_server/tools.ts#L1635) |
+| `conformance_check` | `ro` | stdio | [`consumer_tool_catalog.json:335`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L335) | [`tools.ts:1649`](../../src/scripts/mcp_server/tools.ts#L1649) |
+| `telemetry_report` | `ro` | stdio | [`consumer_tool_catalog.json:342`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L342) | [`tools.ts:1663`](../../src/scripts/mcp_server/tools.ts#L1663) |
+| `catalog_search` | `ro` | _(stub)_ | [`consumer_tool_catalog.json:355`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L355) | _stub-only_ |
+| `catalog_inspect` | `ro` | _(stub)_ | [`consumer_tool_catalog.json:372`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L372) | _stub-only_ |
+| `catalog_load` | `ro` | _(stub)_ | [`consumer_tool_catalog.json:387`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L387) | _stub-only_ |
+| `council_estimate` | `ro` | stdio | [`consumer_tool_catalog.json:401`](../../src/scripts/mcp_server/consumer_tool_catalog.json#L401) | [`tools.ts:1685`](../../src/scripts/mcp_server/tools.ts#L1685) |
 
 ## Glossary
 

package/docs/contracts/tier-3-contrib-plugin.md

@@ -16,7 +16,7 @@ promotion path from Tier-3 → Tier-2, and the non-implementation
 guarantee. Does **not** ship code: `agents/manifests/contrib/` is
 unmanifested until a user asks for an entry by name.
 
-Last refreshed: 2026-05-12.
+Last refreshed: 2026-07-16.
 
 ## Tier definitions
 
@@ -35,8 +35,16 @@ Tier-3 is the explicit overflow bucket.
 The following surfaces are Tier-3 candidates as of 2026-05-12. They
 were surfaced during the
 [`2026-05-12-installer-expansion`](../../agents/runtime/council/sessions/2026-05-12-installer-expansion/synthesis.md)
-council round and have **no entries** in `_VALID_TOOLS`,
-`USER_SCOPE_PATHS`, `SCOPE_SUPPORT`, or the bash `VALID_TOOLS` set.
+council round and had **no entries** in `_VALID_TOOLS`,
+`USER_SCOPE_PATHS`, `SCOPE_SUPPORT` or the bash `VALID_TOOLS` set at
+proposal time.
+
+**Status update (2026-07-16):** the candidates below have since been
+promoted to shipped hosts — each now has entries in `_VALID_TOOLS`,
+`USER_SCOPE_PATHS` and `SCOPE_SUPPORT`, deployed as the universal
+global skill bundle. The bullets are kept as the frozen proposal-time
+record; their rationales describe the pre-promotion state and no
+longer reflect current shipping status.
 
 - `qoder` — community fork, no public adoption signal
 - `trae` — ByteDance IDE, behind login wall
@@ -44,7 +52,7 @@ council round and have **no entries** in `_VALID_TOOLS`,
 - `codebuddy` — Tencent-internal, no public install path
 - `droid` — proposed CLI agent, alpha
 - `warp` — terminal-with-AI; integration shape unclear (PTY vs file marker)
-- `antigravity` — Google research project, no shipping surface
+- `antigravity` — Google's agentic IDE (public since 2026); shipped as a global bundle host (see status update above)
 
 Inclusion in this list is **not** a commitment to ship. Promotion
 requires a real user asking by name (issue, PR, council session, or

package/docs/featured-skills.md

@@ -1,6 +1,6 @@
 # Featured Skills
 
-A curated subset of the 276 skills. Three tiers, one per primary audience. Full catalog lives in [`dist/agent-src/skills/`](../dist/agent-src/skills/); see also [`docs/catalog.md`](catalog.md) for the complete index and [`docs/featured-commands.md`](featured-commands.md) for command-level highlights.
+A curated subset of the 278 skills. Three tiers, one per primary audience. Full catalog lives in [`dist/agent-src/skills/`](../dist/agent-src/skills/); see also [`docs/catalog.md`](catalog.md) for the complete index and [`docs/featured-commands.md`](featured-commands.md) for command-level highlights.
 
 > **Eval-gated messaging note.** Until `task bench --corpus non-dev` reports `selection_accuracy >= 0.60` (see `road-to-product-adoption.md` Phase 1), this page is documentation, not marketing. The entries below are the candidates the corpus tests against; their description quality is what the eval validates.
 
@@ -92,4 +92,4 @@ task lint-featured-skills
 
 CI runs this in `taskfiles/ci-fast.yml`. Stale entries (renamed / removed skill or command) fail the build. See [`scripts/lint_featured_skills.py`](../src/scripts/lint_featured_skills.py) for the matcher.
 
-→ Browse all 276 skills: [`dist/agent-src/skills/`](../dist/agent-src/skills/) · all 189 commands: [`dist/agent-src/commands/`](../dist/agent-src/commands/)
+→ Browse all 278 skills: [`dist/agent-src/skills/`](../dist/agent-src/skills/) · all 190 commands: [`dist/agent-src/commands/`](../dist/agent-src/commands/)

package/docs/getting-started-by-role.md

@@ -2,7 +2,7 @@
 
 > Pick the entry that matches what you do day-to-day. Each section names the three skills you will reach for first and shows whether MCP (no terminal) or CLI (terminal) is the simpler install path for that role.
 
-`agent-config` ships 276 skills, 105 governed rules, and 189 commands. You do not need all of them. Each role below filters to the slice that pays back in week one; the rest stays available and shows up on demand when a task references it.
+`agent-config` ships 278 skills, 105 governed rules, and 190 commands. You do not need all of them. Each role below filters to the slice that pays back in week one; the rest stays available and shows up on demand when a task references it.
 
 > **Taglines at a glance →** [`docs/role-experiences.md`](role-experiences.md) — the one-line pitch per role experience, in one table.
 

package/docs/getting-started.md

@@ -174,7 +174,7 @@ Your agent now understands slash commands:
 | `/fix quality` | Run and fix all quality checks |
 | `/chat-history import` | Pull a prior session into the current chat (resume) |
 
-→ [Browse all 184 active commands](../dist/agent-src/commands/)
+→ [Browse all 190 active commands](../dist/agent-src/commands/)
 
 ---
 

package/docs/governance-advantage.md

@@ -33,7 +33,7 @@ the host lacks (a static host gives it no runtime attention guarantee, but a
 runtime-only system gives it nothing at all here). A runtime-only system contributes *nothing* on these hosts.
 
 ### 2. Intent routing, not a 500-artefact dump
-A prompt does not load all 276 skills. `dist/router.json` (101 routed rules, intent →
+A prompt does not load all 278 skills. `dist/router.json` (101 routed rules, intent →
 skill, keyword/phrase/path triggers) carries the right skill on intent. The
 governance value is *selection discipline* — the host gets the relevant rule
 set, not the whole catalogue as context ballast.

package/docs/proof.md

@@ -27,7 +27,7 @@ evidence pointer, or `task check-claims` fails the build.
 | Claim | Kind | Evidence | Resolves |
 |---|---|---|---|
 | The release process is documented as an inheritable runbook + succession doc, and the project's bus-factor (trailing-90-day distinct human reviewers) is tracked and reported truthfully — currently 1, not implied to be more. | qual | `docs/succession.md#trailing 90 days` | ✅ |
-| 189 commands. | quant | `src/scripts/check_artefact_count_messaging.ts#Artefact-count messaging gate` | ✅ |
+| 190 commands. | quant | `src/scripts/check_artefact_count_messaging.ts#Artefact-count messaging gate` | ✅ |
 | A MEASURED-BUT-NOT-SHIPPED experiment — the thin rule projection reduced eager rule load 78,513 → 13,881 GPT-tokens (whole always-loaded projection 98,529 → 33,897, ~65.6%), but FAILED the quality gate (thin win-rate 36.2% vs required 48%) and does not ship; it un-defers only behind `discipline_profile: essential`. Shipped behavior does NOT include this reduction. Method: `agent-config benchmark` over the pinned token baseline; the baseline is the honest "what the user pays if everything loads eagerly", NOT a synthetic full-corpus strawman (council Q4); quality gate per the Phase-0 paired judge run. | quant | `internal/bench/reports/token-baseline.json#eager_rule_load` | ✅ |
 | The first cross-vendor parity pass (5 orchestration-corpus tasks × 2 vendors × 3 repeats, identical prompts via the council transport, $0.16) measured real per-host finding-count differences — claude-sonnet-4-5 surfaced ~2× the findings of gpt-4o on the multi-file analysis task (median 11 vs 5) while both vendors were identical on the planted hollow-implementation task (2 vs 2) and perfectly silent on the clean-code negative control (0 vs 0, no spurious findings). The per-task `finding_floor` values are calibrated from the cross-host lower envelope and the gate is armed. | quant | `internal/bench/reports/parity-count.json#min over hosts of median` | ✅ |
 | On a weak host (claude-haiku-4-5) the package produces a significant, placebo-controlled discipline lift on scope/downstream traps; on a strong host the same measurement is a published null — the package transplants discipline a weak model lacks, not model intelligence. | quant | `docs/benchmark.md#weak-host-specific` | ✅ |
@@ -40,15 +40,17 @@ evidence pointer, or `task check-claims` fails the build.
 | A hand-rolled, dependency-free BM25 + trigram lexical index resolves the "recalls but does not rank" gap: on the retrieval-precision corpus (9 keyword-overlapping-confuser tasks) it drives the mean top tie-set from 3.333 (the `_score` bucket scorer) to 1.0 — every needed decision uniquely top-ranked — with precision@1 and precision@5 unchanged at 1.0. Method: deterministic, model-free re-ranking of the SAME retrieved entry set; both scorers measured over the identical store via `measure_lexical_ranking.ts`. | quant | `internal/bench/reports/lexical-ranking.json` | ✅ |
 | The whole layer is compiled into host agents with zero runtime daemon. | qual | `docs/contracts/no-runtime-boundary.md#file-first, no-runtime suite` | ✅ |
 | On a 12-fixture option-decision corpus (3 arms × 2 providers, blind rubric judge claude-opus-4-8, pre-registered hypotheses), famous-figure identity framing added nothing beyond the underlying method text (method 5.04 vs figure 4.88, Δ=0.17, sign-test p=0.607), and provider diversity moved judged quality ~15× more than persona identity (provider Δ=2.58 vs identity Δ=0.17); the whole persona layer lifted only +0.08 over bare prompts. Honest null — persona panel-mode stays CUT, evidence-closed. | quant | `internal/bench/reports/persona-placebo.json#honest-null` | ✅ |
+| The only agent layer that publishes the runs where it changed nothing. Deliberately falsifiable — if a reader finds a comparable agent layer publishing its own honest-null benchmark runs, this line updates; that is the point. | comparative | `docs/benchmark.md#honest` | ✅ |
 | On the live end-to-end retrieval benchmark (9 tasks × 3 arms × 3 seeds on claude-haiku, fixture store with keyword-overlapping confusers), the memory retrieval substrate scored precision@5 = 100% (9/9) with 100% poisoned-entry rejection, and the retrieval-on arm passed 27/27 model-scored tasks vs 2/27 with retrieval off and 4/27 with a placebo injection. Known limit stays published: mean tie-set 4.1 means top-k ties break by store order, not relevance (the ADR-116/FTS5 signal). | quant | `internal/bench/reports/second-brain-retrieval.json#retrieval-on` | ✅ |
 | 105 governed rules. | quant | `src/scripts/check_artefact_count_messaging.ts#Artefact-count messaging gate` | ✅ |
 | On a deterministic multi-session recall corpus, the memory substrate produces a measured, placebo-controlled recall lift — memory-on 27/27 vs no-memory 10/27 and vs equal-byte placebo 9/27 (claude-haiku-4-5, n=9 tasks x 3 seeds, sign test p=0.031 for BOTH pairings). Scoped honestly: this is the context-value upper bound (perfect retrieval on a one-fact-per-task corpus), not retrieval precision under a large store. | quant | `internal/bench/reports/second-brain-delta.json` | ✅ |
 | Removing the perfect-retrieval assumption, the substrate's REAL keyword retrieval recalls the needed decision into the top-5 under keyword-overlapping confusers (precision@5 9/9) and the model disambiguates it from the co-injected confusers — retrieval-on 27/27 vs no-memory 5/27 and vs equal-count placebo 5/27 (claude-haiku-4-5, 9 tasks x 3 seeds, sign test p=0.008 both). Named limit: retrieval RECALLS but does not RANK (mean tie-set 3.3, ties broken by store order, not relevance) — the discrimination gap that motivates the SQLite-FTS5 activation path (ADR-116) at larger scale. | quant | `internal/bench/reports/second-brain-retrieval.json` | ✅ |
 | Every artifact the package ships — source AND the condensed projection that reaches consumers — is machine-scanned in CI for hidden-Unicode, mixed-script-confusable, and instruction-smuggling payloads (the rules-file-backdoor class); a finding blocks the release before `npm publish`, not just the merge. | qual | `.github/workflows/publish-npm.yml#lint_agent_security` | ✅ |
-| 276 skills. | quant | `src/scripts/check_artefact_count_messaging.ts#Artefact-count messaging gate` | ✅ |
+| 278 skills. | quant | `src/scripts/check_artefact_count_messaging.ts#Artefact-count messaging gate` | ✅ |
 | Removes only its own keys from a shared host config (matched by JSON-pointer + SHA-256), never a neighbour tool's entries. | qual | `docs/contracts/install-layout.md#JSON-pointer` | ✅ |
+| On the A3 orchestration eval (deterministic verify, measured token deltas), the production-validator subagent returned the correct verdict on both fixtures — NOT READY with the exact `file:line` citation on a planted hollow implementation, READY with zero spurious findings on the clean control — while consuming ~45k fewer tokens than the inline-host baseline on each task. Scope: two planted fixtures on a Claude Code host, not a broad hit-rate. | quant | `internal/bench/orchestration/pv-a3-results.md#token_delta_provenance: measured` | ✅ |
 
-**21 backed claim(s)** — all evidence pointers resolve in CI.
+**23 backed claim(s)** — all evidence pointers resolve in CI.
 
 Artefact counts in public prose (skills, commands, governed rules,
 guidelines, personas) are **generated from source and CI-drift-checked**:
@@ -77,11 +79,11 @@ shipping theater; the full per-cell data is committed at
 `internal/bench/reports/persona-placebo.json`.
 
 **Behavioural-eval coverage — the honest baseline.** Skill *quality* is only
-as good as its measurement. Today **42 of 276** skills carry a behavioural
+as good as its measurement. Today **42 of 278** skills carry a behavioural
 `evals.json`; the highest-traffic / highest-cost tiers (default-surface +
 `rich` + routers) are **fully covered (35 of 35)**, the long tail
-(7 of 241) is not. We publish that gap rather than imply
-"276 evaluated skills": coverage is measured per tier
+(7 of 243) is not. We publish that gap rather than imply
+"278 evaluated skills": coverage is measured per tier
 (`./scripts-run src/scripts/skill_eval_coverage`), **CI-ratcheted so it can
 only rise**, and the priority tiers carry a hard **tier floor**: every
 rich / default-surface / router skill MUST have a behavioural eval or an
@@ -152,6 +154,8 @@ that (`task check-comparison`), so this column can never lie.
 | Capability/benchmark results are published — including the runs where the package changed nothing. | [`docs/benchmark.md`](../docs/benchmark.md) | A category headline figure (e.g. an '84.8%' score) appears across marketing surfaces with no reproducible methodology published — so it cannot be verified either way. | ✅ |
 | Every public claim binds to machine-checked evidence. | [`docs/CLAIMS.md`](../docs/CLAIMS.md) | Marketing claims in this category are not bound to a machine-checked ledger a reader can reproduce. | ✅ |
 | Skills publish their own known limits, each with a witness test. | [`tests/scripts/witness/check_refs_relative_gap.test.ts`](../tests/scripts/witness/check_refs_relative_gap.test.ts) | Published known-limitation surfaces backed by reproducing tests are not a standard artifact in this category. | ✅ |
+| A 30-second single-file wedge exists — one read-only subagent, one command, verdict with file:line — and its promise is a scoped, published eval, not a feature list. | [`internal/bench/orchestration/pv-a3-results.md#token_delta_provenance: measured`](../internal/bench/orchestration/pv-a3-results.md) | Category entry points typically install the full platform (profiles, packs, runtime) before the first felt win; single-artifact entries with a published eval behind the promise are not a standard artifact. | ✅ |
+| The persona-identity question was tested and published as an honest null (identity swap ~zero effect, provider choice ~15x larger) instead of being shipped as theater. | [`internal/bench/reports/persona-placebo.json#honest-null`](../internal/bench/reports/persona-placebo.json) | Named-persona prompts are shipped as a feature across the category without a published A/B separating identity from method. | ✅ |
 
 ## 5. Verify it yourself
 

package/docs/self-review-gate.md

@@ -37,6 +37,20 @@ no-op** (never a failing check) — exactly the `cross-model-canary.yml` pattern
   `enforce: false` (advisory always returns `0` and reports the would-block
   set).
 
+## Escalation on large / claim-affecting diffs
+
+The two in-session lenses are a floor. A **large** diff (≥ 400 changed lines
+across reviewable files) or one that touches a **claim-affecting surface**
+(`docs/CLAIMS.md`, `docs/proof.md`, `docs/comparison.yaml`, or `README.md`)
+warrants the full `ai-council` advisor panel — a spend-bearing multi-model run.
+
+Per blocker `self-review-gate-cost`, the paid council stays governed by the
+standing spend-authorization discipline **at run time**. So the gate does not
+fire council calls itself: `escalationReasons(files, changedLines)` (pure,
+unit-tested) DETECTS the condition, the dry-run plan prints it, and the posted
+review RECOMMENDS a maintainer `/council:pr` run. Detection is deterministic
+and zero-spend; the multi-model run is the maintainer's run-time act.
+
 ## Arming it (maintainer, one flip)
 
 1. Add the `ANTHROPIC_API_KEY` repo secret (per-PR budget sign-off) — turns

package/docs/setup/per-ide/antigravity.md

@@ -2,8 +2,9 @@
 
 Antigravity (Google's agentic IDE) reads the Anthropic-shaped
 markdown skill bundle from its user-scope anchor `~/.agents/`. The
-package deploys via the universal skill convention; project-scope
-bridge is not yet wired (Phase 2.4 anchor).
+package deploys the universal skill bundle at global scope, the same
+projection its other bundle hosts get (codex, continue, roocode and
+the rest) — global-only by design, not a pending gap.
 
 ## Prerequisites
 
@@ -25,7 +26,11 @@ Populates:
 - `~/.agents/personas/` — review-lens personas
 
 (Project-scope `--tools=antigravity` is rejected with exit code 1
-— Antigravity has no documented project-discovery convention yet.)
+— every bundle host is global-only; the package writes the shared
+bundle to a user-scope anchor, not into the project. Antigravity does
+have a native project `.agents/` convention (`skills/`, `agents.md`,
+`workflows/`), but a project-scope projection is not part of the
+shipped surface.)
 
 ## How to use
 

package/docs/us-vs-the-category.md

@@ -0,0 +1,38 @@
+# Us vs. the category — the honest version
+
+<!-- claim:positioning-honest-nulls -->**The only agent layer that publishes
+the runs where it changed nothing.**<!-- /claim --> That line is deliberately
+falsifiable: find a comparable agent layer publishing its own honest-null
+benchmark runs and it updates — that is the point. The category's growth
+playbook is headline numbers nobody can re-run; ours is a claims ledger CI
+refuses to let drift.
+
+This page is the adoption-context frame around the CI-locked comparison table.
+The table itself lives on the proof page — every "our evidence" pointer is
+resolved by `task check-comparison` on every build, the category column only
+ever describes what is publicly observable, and no competitor is named or
+counter-claimed:
+
+**→ [The comparison table, CI-checked on every build](proof.md#4-what-is-checkable--us-vs-the-category)**
+
+## What the rows mean for an adopter
+
+- **No runtime.** Nothing to keep alive, nothing that writes state behind
+  you — the whole layer compiles into your host agent's own config surface.
+- **Published nulls.** The benchmark page keeps the runs where the package
+  changed *nothing* — including the persona-theater question: swapping the
+  advisor identities was measured at ~zero effect while the provider choice
+  mattered ~15× more, so persona identity was published as a placebo
+  (<!-- claim:persona-identity-placebo-null -->the persona-identity A/B is a
+  backed honest null<!-- /claim -->) instead of shipped as a feature. A
+  council implementation with falsifiable verdicts — including about itself.
+- **A wedge, not a funnel trap.** The 30-second entry is one read-only
+  subagent with a scoped, published eval behind its promise
+  ([what it catches](wedge/production-validator/README.md)) — the platform is
+  the second date, not the first.
+
+## Verify it yourself
+
+Every command on the [proof page § Verify it yourself](proof.md#5-verify-it-yourself)
+runs on a fresh checkout. If a claim here does not resolve, CI is red before
+you ever read it.

package/docs/wedge/production-validator/README.md

@@ -7,6 +7,14 @@ system — not just green tests over hollow code.
 
 One file. No install of anything else. No runtime, no daemon, no account.
 
+<!-- claim:wedge-hollow-detection -->On its published eval it returned the
+correct verdict on both fixtures — `NOT READY` with the exact `file:line` on a
+planted hollow implementation, `READY` with zero spurious findings on the clean
+control — at ~45k fewer tokens per task than the same check run inline
+([evidence](../../../internal/bench/orchestration/pv-a3-results.md)). Two
+planted fixtures on a Claude Code host — a scoped, reproducible promise, not a
+hit-rate.<!-- /claim -->
+
 ## What it does
 
 Before you call a feature done, `@production-validator` greps the shipped path
@@ -46,6 +54,22 @@ production-validator
   Final gate: NOT READY — refund() is a stub; validate against the real provider.
 ```
 
+## Optional: record your first run (local-only, opt-in)
+
+Nothing is measured unless you run it. After your first verdict, you can keep a
+private local record — one aggregate line, written into YOUR repo, no network:
+
+```bash
+bash <(curl -fsSL https://raw.githubusercontent.com/event4u-app/agent-config/main/docs/wedge/production-validator/first-run-check.sh) ready
+```
+
+Outcomes: `ready` · `not-ready` · `abandoned`. The log
+(`.claude/wedge-first-run.local.log`) never leaves your machine — the package
+runs zero telemetry by default
+([contract](../../contracts/adoption-signal-floor.md)). The same outcome
+vocabulary feeds the proctored install-friction study
+(`agents/recruit-sessions/_install-friction-runbook.md`).
+
 ## Why this one
 
 It carries an anti-hallucination discipline (Evidence-First): "done" is a claim

package/docs/wedge/production-validator/first-run-check.sh

@@ -0,0 +1,42 @@
+#!/usr/bin/env bash
+# Opt-in, local-only first-run check for the production-validator wedge.
+#
+# Default-off by construction: nothing runs unless YOU invoke this script.
+# It writes ONE aggregate line to a local log in your repo and never opens a
+# network connection — consistent with the package's no-telemetry posture
+# (docs/contracts/adoption-signal-floor.md). Delete the log any time.
+#
+# Usage (after your first @production-validator verdict):
+#   bash first-run-check.sh ready       # first verdict was READY
+#   bash first-run-check.sh not-ready   # first verdict was NOT READY
+#   bash first-run-check.sh abandoned   # you gave up before a verdict
+#
+# The same outcome vocabulary feeds the B9 install-friction study
+# (agents/recruit-sessions/_install-friction-runbook.md) when a session is
+# proctored; unproctored users simply keep a private local record.
+set -euo pipefail
+
+OUTCOME="${1:-}"
+case "$OUTCOME" in
+  ready|not-ready|abandoned) ;;
+  *)
+    echo "usage: bash first-run-check.sh <ready|not-ready|abandoned>" >&2
+    exit 2
+    ;;
+esac
+
+AGENT_FILE=".claude/agents/production-validator.md"
+LOG_FILE=".claude/wedge-first-run.local.log"
+
+if [ ! -f "$AGENT_FILE" ]; then
+  echo "NOT INSTALLED — $AGENT_FILE missing. Install first:" >&2
+  echo "  mkdir -p .claude/agents && curl -fsSL https://raw.githubusercontent.com/event4u-app/agent-config/main/docs/wedge/production-validator/production-validator.md -o $AGENT_FILE" >&2
+  exit 1
+fi
+
+# One aggregate line: date + outcome. No user, host, repo, or content data.
+printf '%s wedge=production-validator first_run=%s\n' \
+  "$(date -u +%Y-%m-%dT%H:%M:%SZ)" "$OUTCOME" >> "$LOG_FILE"
+
+echo "Recorded locally in $LOG_FILE (never leaves this machine)."
+echo "Add it to .gitignore if you don't want it committed."

package/package.json

@@ -1,6 +1,6 @@
 {
     "name": "@event4u/agent-config",
-    "version": "9.3.0",
+    "version": "9.4.0",
     "description": "Universal AI Agent OS — audited skills, governance rules, commands, and templates for AI coding tools (Claude Code, Cursor, Windsurf, Copilot).",
     "license": "MIT",
     "private": false,

package/src/scripts/_lib/catalog_score.ts

@@ -0,0 +1,64 @@
+/**
+ * Catalog search scoring — a thin field-boosted wrapper over the shared
+ * `LexicalIndex` (BM25 + trigram prefilter).
+ *
+ * DECISION (road-to-persona-library-harvest.md Phase 2.3): the roadmap sketched
+ * porting the upstream Hermes `_score` (token overlap + name/description
+ * boosts). We do NOT — `_lib/lexical_index.ts` already ships a calibrated BM25
+ * core (ADR-061-sanctioned, dependency-free) with a coverage² multi-term
+ * weighting and a trigram candidate prefilter. Maintaining a second lexical
+ * scorer would be duplication. Instead we reuse `LexicalIndex` and get the
+ * field boost by REPEATING the name (×3) and description (×2) terms in each
+ * doc's indexed text, so a query term hitting the name outranks the same term
+ * buried in a tag. No new scorer.
+ */
+import type { CatalogIndexEntry } from '../build_catalog_index.js';
+import { LexicalIndex } from './lexical_index.js';
+
+export interface CatalogFilter {
+    cls?: CatalogIndexEntry['cls'];
+    /** Match against the entry's tags (packs/domain/tier). */
+    pack?: string;
+    limit?: number;
+}
+
+export interface CatalogHit {
+    entry: CatalogIndexEntry;
+    score: number;
+}
+
+/** Field-boosted searchable text: name ×3, description ×2, tags ×1. */
+function _docText(e: CatalogIndexEntry): string {
+    return [e.name, e.name, e.name, e.description, e.description, ...e.tags].join(' ');
+}
+
+function _filter(entries: readonly CatalogIndexEntry[], f: CatalogFilter): CatalogIndexEntry[] {
+    return entries.filter((e) => {
+        if (f.cls && e.cls !== f.cls) return false;
+        if (f.pack && !e.tags.includes(f.pack)) return false;
+        return true;
+    });
+}
+
+/**
+ * Rank catalog entries against a free-text query. The index is built over the
+ * (filtered) candidate set so IDF reflects the searched scope. Empty or
+ * whitespace query → no hits (a search tool must not dump the catalog).
+ */
+export function searchCatalog(
+    entries: readonly CatalogIndexEntry[],
+    query: string,
+    filter: CatalogFilter = {},
+): CatalogHit[] {
+    const q = query.trim();
+    if (q.length === 0) return [];
+    const candidates = _filter(entries, filter);
+    if (candidates.length === 0) return [];
+
+    const index = new LexicalIndex(candidates.map((e) => ({ id: e.id, text: _docText(e) })));
+    const byId = new Map(candidates.map((e) => [e.id, e]));
+    const limit = filter.limit && filter.limit > 0 ? filter.limit : 8;
+
+    const ranked = index.rank([q]).filter((r) => r.score > 0);
+    return ranked.slice(0, limit).map((r) => ({ entry: byId.get(r.id) as CatalogIndexEntry, score: r.score }));
+}